The effect of FMT and vitamin C on immunity-related genes in antibiotic-induced dysbiosis in mice

GEO dataset

The Gene Expression Omnibus (GEO) database is a gene expression database created and maintained by the National Center for Biotechnology Information NCBI. We selected the data set with the registration number GSE58087 to analyze the effect of antibiotics on the organism. The data set has several samples, and we selected four groups of mice with normal ileum for 8 weeks, low-dose penicillin for 8 weeks ileum, normal week liver for 30 weeks, and low-dose penicillin for 30 weeks liver for differential gene expression analysis and functional enrichment analysis (Fig. 1).

Animals

Fifteen 4-week-old C57BL/6J male mice (15∼25g) were purchased from SPF(Beijing) BIOTECHNOLOGY Co., Ltd. and housed individually in an experimental environment at 23 °C with a 12-hour light/dark cycle. During the experiment, mice had free access to sterilized food and water. After 10 days of acclimation, three mice were used as a control group (NO-ABX) for group N. Twelve mice received antibiotics for 21 days and were later randomly divided into four groups of three mice each in a cage for groups ABX, F, VF, and NS. The ABX group was executed by cervical dislocation after receiving antibiotics for 21 days, and intestinal and liver tissue samples were frozen in liquid nitrogen and stored at −80 °C. The NS group received antibiotics for 21 days followed by saline for 21 days, while the F group received fecal microbial transplantation for 21 days and the VF group received fecal microbial transplantation and vitamin C for 21 days. On day 42 of the experiment, mice in the N, F, VF, and NS groups were executed by cervical dislocation, and intestinal and liver tissue samples were frozen in liquid nitrogen and stored at −80 °C. The collected tissue samples were subjected to quantitative real-time PCR (Fig. 1). All animals were cervically dislocated for sample collection, and no residual animals remained after the experiment.

The Research Ethics Committee of Chongqing University of Posts and Telecommunications approved all experimental procedures (approval reference numbers: 20210324-1).

Antibiotic therapy and fecal microbiota transplantation

Single antibiotics have limited clearance efficiency for complex microbial ecosystems (Gopalakrishnan et al., 2021). Christopher Staley’s study showed that an extended course of antibiotics better disrupted the original intestinal microbiota in mice (Staley et al., 2017), while Vancheswaran Gopalakrishnan found that a 3-week antibiotic regimen better promoted microbial community transformation and better acceptance of donor flora (Gopalakrishnan et al., 2021). After ten days of self-adaptation, we selected neomycin (0.5 mg/ml), which has a small degree of systemic absorption, and ampicillin (1 mg/ml), which has a large degree of systemic absorption, for antibiotic treatment of mice (Lynn et al., 2021; Staley et al., 2017). It was dissolved in sterile drinking water and filtered, obtained ad libitum by the mice, and changed once a day for 21 days of administration.

The role of FMT is also limited in that it allows the establishment of only a small fraction of the human bacterial community (Wos-Oxley et al., 2012). Korry J Hintze suggested that two to three weeks after antibiotic treatment is the best time for fecal microbiota transplantation (Hintze et al., 2014), and for better retention of the donor mouse flora in the recipient mice, we chose to transplant mice with fecal microbiota for 21 consecutive days. Fecal microbiota transplantation (FMT) was performed using samples made from untreated feces of healthy mice. The feces of healthy mice were mixed with sterile saline (40 mg/ml) and homogenized. After homogenization, centrifugation (800g, 3min, 4 °C) was performed and the supernatant was collected, which was the transplantation sample. The treated mice were given 100uL of supernatant by gavage for 21 consecutive days, and an additional 100ul of 200mg/kg VC solution was given to mice in group 4. Group 2 mice recovered naturally and were given only 200ul of sterile saline, and normal mice in Group 5 were given sterile food and quoted water as normal (Fig. 1).

Quantitative Real-time PCR

Total RNA was extracted from collected intestinal and liver tissues using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer’s protocol, and the purity of RNA was measured using spectrophotometric analysis (Quawell Q3000, USA). Next, RNA was reverse transcribed to cDNA using the RevertAid First Strand cDNA Synthesis Kit (Thermo Scientific, Waltham, MA, USA). The quantitative real-time PCR (qRT-PCR) analysis was performed using SYBRPRIME qPCR Kit (Baoguang Biotechnology Co., Ltd., Chengdu, China) and Bio-Rad iQ5 software (Bio-Rad, Hercules, CA, USA). GAPDH was analyzed in each sample to normalize expression. Three biological replicates were performed for each gene, and each replicate contained three technical replicates. The primers used in this study are listed in Table 1. relative expression was analyzed by the 2-ΔΔCt method (Livak method). The data for this study can be obtained by downloading the Supplementary File.

Statistical analysis

The experimental data were expressed as mean ±standard deviation, differences between the two groups were analyzed using t-test, and differences between the three groups were analyzed using analysis of variance (ANOVA). GEO data were transformed from ENTRZ ID annotation to SYMBOL ID by R language, and after voom normalization by limma (Ritchie et al., 2015), differential genes were screened and GO enrichment analysis and KEGG pathway enrichment analysis were performed using clusterProfiler (Yu et al., 2012). Differentially expressed mRNA clustering analysis was performed with FC values using the heatmap function of R. Based on the ploidy change log(FC) ≥ 1.0 and P.value < 0.05 as the differential screening threshold, and based on this, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed.GO analysis included biological processes, cellular components, and molecular functions; KEGG pathway analysis inferred the biological signaling pathways they might be involved in, with p .value< 0.05 as the threshold for significant enrichment.

Low-dose penicillin affects the immune pathway of the Ileum in mice

To investigate the effect of antibiotics on intestinal gene expression, the GSE58087 microarray dataset was selected for analysis. Four male mice were exposed to low-dose penicillin (LDP) (6.7 mg/L) from birth and three mice were born normally as control, ileum was collected at eight weeks of age, RNA was extracted, and the differences were analyzed, with —FC—>1 representing significant and P value < 0.05 representing statistically significant screening for differential genes. As shown in the volcano plot in Fig. 2A, a total of 35 up-regulated genes vs. 12 down-regulated genes, and was plotted by hierarchical clustering (see Fig. 2B). Each column represents the expression pattern of a sample; high and low expression levels are indicated by red and green lines, respectively, with a clear distribution of differential genes. GO enrichment analysis of differential genes, in biological processes, as in Fig. 2C, is mainly involved in antibody processing and presentation, interferon IFN-γ response, exogenous peptide antibody processing and presentation, and peptide antibody processing and presentation via MHC-II. In the molecular function, as in Fig. 2D, the peptide and amide binding, MHC-II protein complex binding, and antibody and peptide antibody binding processes are mainly carried out. Antigen presentation as a bridge between natural and acquired immunity. Antigen processing is the process by which cells in vivo collect antigens and degrade them into peptides, while antigen presentation is the process by which processed antigen peptides are presented to the cell surface for easy recognition by T cells. MHC can bind to peptides to form protein complexes and provide feedback to T cells by presenting antigenic peptides. MHC-II mainly presents exogenous peptides (i.e., from extracellular sources), is recognized by CD4+ T cells (helper T cells), is expressed only on the surface of antigen-presenting cells such as macrophages, dendritic cells, and B cells, and is known to be involved primarily in the immune response. Low-dose penicillin alters the expression of immune-related genes in mice and disrupts the original balance of immune gene expression in the mouse ileum. A KEGG pathway enrichment analysis was then performed, as shown in Fig. 2E. It was mainly enriched to antibody processing and presentation, hematopoietic cell lineage, allogeneic rejection, graft-versus-host disease, and TH1 and TH2 cell differentiation pathways, highly suggesting that it is closely related to the immune system. This again demonstrates the involvement of the intestinal microbiota in numerous immune-like pathways.

Three other normally grown mice and three mice receiving low penicillin injections had their livers collected at 30 weeks of age and RNA extracted. differential analysis revealed three upregulated genes, Neat1, Cirbp, and Rian, and five downregulated genes, Hspa1b, Ugt2b38, Hspb1, Cidec, and Slc10a2 (Table 2). Enrichment analysis of differential genes, in biological processes, as in Fig. 3A, were mainly involved in chaperone-mediated protein folding, response to unfolded proteins, response to topologically incorrect proteins, protein folding, response to temperature stimulation, negative regulation of cell activation response, associated with protein folding, temperature stimulation, and cell activation. In the molecular functions, as in Fig. 3B, the protein folding chaperones were mainly involved. Then KEGG pathway enrichment analysis was performed, as in Fig. 3C, enriching the pathway of bile secretion.

Identification of 12 key genes

Through literature searches and related websites, 12 key genes in the differential genes were selected for validation: Cd74, H2-Aa, Ccl5, H2-DMa, Psmb9, Psmb8, Cd3g, Ubd, Gzmb, Saa2, Saa1, Saa3, genes mostly enriched in signaling pathways of immunity and inflammation (Table 3). CD74 is an invariant chain expressed on the plasma membrane and presents an evolutionarily conserved type II membrane protein, which is involved in many different activities and is a key gene in both immune and inflammatory signaling pathways (Borghese & Clanchy, 2011). Major histocompatibility complex class II (MHC II) is an important immune regulatory molecule that plays an important role in both antigen presentation and immune cell development. H2-Aa, as part of the MHC class II protein complex, has been associated with a variety of diseases, including autoimmune diseases and cancers of the gastrointestinal system, and is a significant pro-inflammatory-related gene (Zhao et al., 2022). Ccl5 is homologous to human CCL5 (C-C motif chemokine ligand 5), a key pro-inflammatory chemokine that induces in vitro migration and recruitment of a variety of cells, including T cells, and plays a role in a variety of biological processes including cancer and atherosclerosis. ccl5 also interacts with a variety of other receptors to regulate cellular activity, receptor and chemokine levels and maintain homeostasis in vivo (Lapteva & Huang, 2010). H2-DMa facilitates the exchange of peptides bound to MHC class II molecules and plays an important role in humoral immune responses, but different class II molecules have different sensitivities to it (Alfonso et al., 2001). Psmb9 and Psmb8 are genes encoding immunoproteasome subunits that act upstream or within antigen processing, bacterial presentation, and response, and have been implicated in both lymphoma and renal cell carcinoma. CD3G is involved in the formation of the TCR-CD3 complex and immune response, and its mutation often leads to diseases such as inflammation and cancer (Birnbaum et al., 2014). Ubiquitin D (UBD) is a member of the ubiquitin-like (UBL) modifier family, highly expressed in a variety of cancers including colorectal cancer (CRC) (Su et al., 2021), and is also involved in cellular protein metabolic processes; positive regulation of I-kappaB kinase/NF-kappaB signaling; and positive regulation of apoptotic processes. It acts upstream or internally in response to interferon-gamma and response to tumor necrosis factor, located in the cytoplasm and nucleus and expressed in the digestive system, aorta, male gonads or organs, olfactory lobes, and thymus. GZMB encodes a member of the granzyme subfamily of proteins and is part of the S1 family of serine proteases. It has a positive role in the clearance of tumor cells, bacteria, parasites, and viruses, as well as in the recognition, diagnosis, and treatment of diseases, but have a negative role in extracellular functions (Wang et al., 2021). The acute phase members of the mouse serum amyloid A (SAA) family, Saa1, Saa2, and Saa3, are highly similar at both the nucleotide and protein sequence levels, but respond differently to different pro-inflammatory cytokines (Thorn & Whitehead, 2002), and SAA3 was found to be pro-atherogenic (Thompson et al., 2018).

Differential expression of key genes in the intestine and liver after antibiotic treatment

To verify whether different antibiotics would affect gene expression, we treated mice with neomycin and ampicillin for 21 days, then recovered naturally for 21 days, and then examined their relative expression to the normal group by qPCR. Mice treated with antibiotics for 21 days showed significantly elevated expression of 11 genes in the intestine, except for H2-DMa, and most of them were significantly different. After 21 days of normal saline feeding, i.e., natural recovery, GZMB, CD3G, PSMB8, SAA3, H2-DMA, and SAA1 all returned to about normal levels, but H2-AA, PSMB9, CD74, UBD, SAA2, and CCL5 still had expression differences, with H2-AA, PSMB9, and CCL5 showing reduced expression and CD74, UBD, and SAA2 expression was elevated (Fig. 4, Tables S1, S2). The same expression trend remained in the liver, with significantly higher expression except for PSMB9, but the increase was not as high as in the intestine, and most genes also returned to normal levels after natural recovery, with only H2-DMA, SAA2, CCL5, and PSMB9 remaining significantly lower (Fig. 5, Tables S3, S4).

Fecal microbiota transplantation in healthy mice may have additional effects on key gene expression

To determine whether fecal microbiota transplantation after antibiotic treatment facilitates the re-establishment of intestinal microbiota and the effect on intestinal versus liver genes, we continued FMT for 21 days in experimental groups of mice treated with antibiotics for 21 consecutive days. The qPCR results showed that FMT did not help to restore genes to normal levels in intestinal or liver tissues, but in intestinal tissues, genes that were restored more normally in the NS group (also known as the natural recovery group) were highly expressed in the FMT group, with only SAA2 significantly lowly expressed, CD74 still highly expressed, and UBD largely unexpressed. For liver tissues, the expression of H2-DMA remained low, the expression of SAA1, SAA2, SAA3, and PSMB8 was significantly higher, H2-AA was significantly lower, and only GZMB, CD3G, and PSMB9 changed in a good direction (Fig. 6, Table S5).

Vitamin C contributes to the restoration of gene expression in intestinal and liver tissues

Studies have shown that specific dietary interventions and micronutrient modulation of intestinal microbes are positive therapeutic strategies, while vitamin C (ascorbic acid) has been shown to have health benefits and to be involved in biological processes that support the immune system (Carr & Maggini, 2017). Vitamin C supplementation also has a protective effect against oxidative stress and has received attention for its role in oxidative stress-related pathological diseases. Studies have confirmed that supplementation with high doses of vitamin C can manipulate the composition of the intestinal microbiota, thereby altering the intestinal microbiota (Otten et al., 2021).

To investigate whether vitamin C could affect fecal microbiota transplantation, we designed an experimental group in which mice were transplanted with fecal microbes for 21 consecutive days after 21 days of Abx treatment and given the same dose of vitamin C. The qPCR results showed that for intestinal tissues, GZMB, CD3G, H2-AA, PSMB9, SAA1 had better results and were closer to the genes in the normal group, reducing the expression of their highly expressed genes in the F group, these genes function in Granzyme Pathway, Intrinsic Pathway for Apoptosis, Programmed Cell Death, Immune response NFAT in immune response, TCR Signaling (Qiagen), Class I MHC mediated antigen processing and presentation, Cytokine Signaling in Immune system, among other pathways. The addition of vitamin C had no significant effect on CD74, UBD, PSMB8, SAA3, H2-DMA, SAA2, and CCL5, but it did not cause worse effects either. In liver tissues, more favorable effects were found for GZMB, CD3G, H2-AA, PSMB8, SAA1, and less favorable effects for the other seven genes. It is also important to note that the expression of genes in both tissues was higher in the intestine than in the liver for group F genes, and the same was true after the addition of vitamin C. This suggests that vitamin C acts directly in the intestine and more weakly in the liver. In conclusion, for intestinal tissues, the introduction of vitamin C restored the genes more to normal levels, and for liver tissues, vitamin C mostly reduced the expression of highly expressed genes (Fig. 7, Table S5).